Remote domes!
White knuckles
It seems that everyone is getting into remote imaging these days. If you have an automated rig in a dome, you don't actually need to be present to make it work. I mean, what's the difference between sitting at a computer two metres from your rig and sitting at a computer 200 kilometres from your rig? The software is the same, the images are the same.
You know I'm going to tell you what's different. Security.
Imagine something goes wrong. Suddenly your computer screen goes blank. You don't know if the power has gone out, and if it hasn't, you don't know if your mount is about to mash your expensive camera and filter wheel against the pier. Worse, the dome is open to the elements and there might be a rain storm coming.
There's really no alternative but to drag yourself to the car and make a two hour journey out to your remote site.
But astrophotographers do it.
My state's astronomical society is piloting a couple of locations. A few years ago, the society set up a 3.5 metre Sirius dome at its dark sky site in Central Victoria, and we astrophotographers could see (or rather hear) the dome going through its routines. It's set off to one side of the astrophotography field, so you'd hear it opening in the dark.
I don't know about the others there, but I'd quietly admire the people who had the knowledge and access to be able to do all that. And from Melbourne!
Security concerns are manageable
It's not cheap, but you can make the whole facility work reliably and securely.
First, you need a UPS - just enough to give you a few minutes of power in the case of an outage. If you're on mains power, the ScopeDome has a sensor that you simply plug into the unprotected mains outlet. If that goes out, the first priority is to shut the dome.
Second, you need a weather station. We use Lunatico Cloudwatchers. This keeps an eye on sky temperature (a cold sky indicates no clouds), wind, rain, etc. Voyager (the central software) looks at the data from the Cloudwatcher every 30 seconds or so, and so at the first hint of danger it shuts the dome and suspends whatever is going on. Depending on how sophisticated your scripting is, you can resume when conditions improve.
Third, you need eyes on the telescope. We use an IR security camera that hooks into the network. Sometimes these can sit on the floor of the dome and look at the telescope, showing you what it's doing - slewing when it's meant to, flipping when it's meant to, etc., and you can keep an eye on the cabling. Preferably, though, you mount the camera on the rotating part of the dome opposite the dome's shutter. That way you can have a reasonable idea as to how well the dome is slaving.
Cut to present
The society has installed two more of these remote domes - these ones are 2 metre ScopeDomes, supplied by the company I've been working for for the last few years. They're right next to the 3.5m Sirius dome. One has a 100mm Esprit with a colour camera on a CEM40, and the other has an awesome Astroworx 10" f/4 carbon fibre Newtonian and a ToupTek ATR2600M monochrome camera with a filter wheel.
Sidereal Trading has installed quite a few of these domes in sometimes dramatically remote locations. Not only have they gone to clubs, they've gone to schools, at least one university, and quite a few private astronomers. The less usual one was a caravan park where the developer wanted to have a dome for tourists to have an astronomy experience. We installed it but I have yet to hear of it being active. I understand that the developer hasn't been able to get a volunteer astronomer to be a tour guide, and sadly. the project may have been abandoned. I can't do it, the commute is two days' drive.
How does it work?
The dome has electricity and Internet. This can be mains power and NBN Internet, but it can also be solar panels, battery storage and Starlink. Normally, the dome and computer remain on all the time (so get an industrial machine, preferably fanless), and you connect via a "screen scraper" program like Anydesk or Chrome Remote Desktop.
From here, you can turn components (like the camera, focuser, etc.) on and off using a network-enabled relay. If you're really clever, you can just leave that relay on, and remotely control that to turn the computer on. It depends on how sophisticated you want to be.
Once everything is on, you can get into Voyager. This is a program that controls all your components, so you can use it as an "orchestra conductor". The Voyager screen shows you what you're connected to, what and they're all doing. It looks a bit like this:
Voyager can work "on the fly", where you can run the facility like you would if you were there, flitting from one target to the other, or you can use a script to automate everything. If you do this, you can make the whole place work on its own. The way I've set this one up is that I tell it what to shoot, and when to start. If I do this in the afternoon, it does this:
- wait until shortly before "astronomical night"
- turn on the electric components (mount, camera, etc.)
- starts cooling the camera
- finds home on the dome in preparation for slaving
- point the telescope to a point slightly left of the intersection of the meridian and equator (a good place to calibrate some things)
- slave the dome (to keep the opening in front of the telescope)
- open the dome (if it's safe)
- focus the telescope
- plate solve to synchronise the mount to the sky
- calibrate the autoguider
- wait (with the shutter closed) until the target is more than 30° above the horizon
- point to the target and start taking images
- keep on doing this until either the target starts to set or it gets light.
At any time, if the Cloudwatcher spots a cloud, it shuts the dome and waits. If conditions get better it resumes.
But does it work?
It's a constant struggle. Because I'm primarily installing facilities for clients, no one system is identical. There's always different domes, different cameras, different mounts, etc. I'm not really happy with any facility. They all work, but they need to be babied. Our development facility is nearly there, but it has a camera that seems to have a bug in its firmware, meaning the all-night script works flawlessly but has to stop for me to manually chill the camera.
We're close. That's all I can say.
I've taken some really nice test shots though. Here's one. This is from a school dome with a 14" f/4.5 Newtonian. The mirror needs cleaning.



Comments
Post a Comment